The invention relates to determining a point-of-interest (POI) that accurately indicates the actual location of a user based on various types of information, such as user activity contexts, environment contexts, social contexts, and crowd-sourced data.
Many times, people do not specifically keep track of and/or record the various places they visit. To alleviate this problem, some location-based social applications allow users to record and share the various places and physical locations that they visit in real-time. For example, these applications may be configured to record and/or track the various places that the users visit over a particular period of time, and allow the users to “check-in” to certain places that they want to share with friends and the public as their points or places of interest.
However, these location-based applications are limited and deficient in various ways. For instance, the POIs recorded by the applications may not accurately indicate the actual location of the users, especially if there are other POIs within the same geographical vicinity, such as a shopping mall, and/or if the users park their vehicles in a location relatively far away from the actual POI and subsequently moves from the parking location to another location.
In that regard, there is a need to automatically determine the user's actual POI based on various types of contexts.